ORIGINAL RESEARCHpublished: 09 January 2019

doi: 10.3389/fcvm.2018.00189

Frontiers in Cardiovascular Medicine | www.frontiersin.org 1 January 2019 | Volume 5 | Article 189

Edited by:

Hyung J. Chun,

Yale University, United States

Reviewed by:

Stephanie Thorn,

Yale University, United States

Sudarshan Rajagopal,

Duke University Health System,

United States

*Correspondence:

Naomi C. Chesler

naomi.chesler@wisc.edu

Specialty section:

This article was submitted to

Heart Failure and Transplantation,

a section of the journal

Frontiers in Cardiovascular Medicine

Received: 23 October 2018

Accepted: 13 December 2018

Published: 09 January 2019

Citation:

Mulchrone A, Kellihan HB,

Forouzan O, Hacker TA, Bates ML,

Francois CJ and Chesler NC (2019) A

Large Animal Model of Right

Ventricular Failure due to Chronic

Thromboembolic Pulmonary

Hypertension: A Focus on Function.

Front. Cardiovasc. Med. 5:189.

doi: 10.3389/fcvm.2018.00189

A Large Animal Model of RightVentricular Failure due to ChronicThromboembolic PulmonaryHypertension: A Focus on FunctionAshley Mulchrone 1, Heidi B. Kellihan 2, Omid Forouzan 1, Timothy A. Hacker 3,

Melissa L. Bates 4,5, Christopher J. Francois 6 and Naomi C. Chesler 1,3*

1Department of Biomedical Engineering, Univeristy of Wisconsin-Madison, Madison, WI, United States, 2 School of Veterinary

Medicine, University of Wisconsin-Madison, Madison, WI, United States, 3Department of Medicine, University of

Wisconsin-Madison, Madison, WI, United States, 4Department of Health and Human Physiology, University of Iowa, Iowa

City, IA, United States, 5Department of Pediatrics, University of Iowa, Iowa City, IA, United States, 6Department of Radiology,

University of Wisconsin-Madison, Madison, WI, United States

Chronic thromboembolic pulmonary hypertension (CTEPH) is a debilitating disease that

progresses to right ventricular (RV) failure and death if left untreated. Little is known

regarding the progression of RV failure in this disease, greatly limiting effective prognoses,

and therapeutic interventions. Large animal models enable the use of clinical techniques

and technologies to assess progression and diagnose failure, but the existing large animal

models of CTEPH have not been shown to replicate the functional consequences of the

RV, i.e., RV failure. Here, we created a canine embolization model of CTEPH utilizing

only microsphere injections, and we used a combination of right heart catheterization

(RHC), echocardiography (echo), and magnetic resonance imaging (MRI) to quantify RV

function. Over the course of several months, CTEPH led to a 6-fold increase in pulmonary

vascular resistance (PVR) in four adult, male beagles. As evidenced by decreased cardiac

index (0.12± 0.01 v. 0.07± 0.01 [L/(min∗kg)]; p < 0.05), ejection fraction (0.48± 0.02 v.

0.31 ± 0.02; p < 0.05), and ventricular-vascular coupling ratio (0.95 ± 0.09 v. 0.45 ±

0.05; p < 0.05), as well as decreased tricuspid annular plane systolic excursion (TAPSE)

(1.37 ± 0.06 v. 0.86 ± 0.05 [cm]; p < 0.05) and increased end-diastolic volume index

(2.73 ± 0.06 v. 2.98 ± 0.02 [mL/kg]; p < 0.05), the model caused RV failure. The ability

of this large animal CTEPH model to replicate the hemodynamic consequences of the

human disease suggests that it could be utilized for future studies to gain insight into the

pathophysiology of CTEPH development, following further optimization.

Keywords: pulmonary embolization, pulmonary hemodynamics, right ventricular afterload, effective arterial

elastance (Ea), pulmonary vascular resistance (PVR)

INTRODUCTION

Chronic thromboembolic pulmonary hypertension (CTEPH) is a debilitating, fast progressingvascular disease associated with poor prognosis and significant morbidity and mortality (1–3). It isone of the most common and potentially curable subsets of precapillary pulmonary hypertension(PH) (4, 5) and is characterized by the obstruction of the pulmonary vasculature from unresolved,

Mulchrone et al. CTEPH: A Focus on Function

organized thromboemboli. The diagnosis of CTEPH is madeby mean pulmonary artery pressures (mPAP) ≥25 mmHg,pulmonary capillary wedge pressures (PCWP) ≤15 mmHg atrest, and evidence of thromboemboli by an imaging modality(5, 6). Since patients are often asymptomatic or misdiagnosed,CTEPH is typically advanced at the time of diagnosis(7–9).

The mechanical obstruction of the pulmonary vascular bed inCTEPH increases pulmonary vascular resistance (PVR) and rightventricular (RV) afterload. The RV can adapt to the increasedafterload for some time to maintain cardiac output (CO), butwithout treatment, CO and ejection fraction (EF) typically dropand death ensues (2, 10, 11). There are several critical knowledgegaps in this process, including the mechanistic transition fromadaptation to maladaptive remodeling, the functional precursorsof failure, and the biological indicators of the failed RV. Theseknowledge gaps limit the development of effective prognoses andoptimized patient care (6).

A common approach to address pathophysiologicalknowledge gaps is preclinical or animal models of disease.Due to their cost-effectiveness, efficiency, and potential forgenetic manipulation, rodents are frequently used. Commontechniques for studying venous thrombus generation/resolutionor CTEPH include pulmonary artery (PA) ligations (12),balloon occlusions (13, 14), and microsphere injections (3).These models provide limited insight into clinical practice asthey generally fail to replicate RV failure. Neto-Neves et al.created a successful CTEPH rat model utilizing microsphereinjections in conjunction with a vascular endothelial growthfactor (VEGF) receptor tyrosine kinase inhibitor (SU5416;Tocris Bioscience, Bristol, UK) that did demonstrate RVremodeling and RV dysfunction, but only one animal wasstudied out to heart failure (15). Moreover, findings fromsmall-animal models can be strain-specific with significantinter- and intra-species variation, and/or have accelerateddisease progression not consistent with clinical presentation(2, 3, 10).

In contrast, large animal models, which incur increased costsand complexities in housing and care, more closely mimichuman physiology and pathophysiology (16). Investigationsin large animals can also utilize the same techniques andtechnologies as clinical studies.While acute embolismmodels arerelatively easy to induce, capturing the hallmark characteristicsof clinical CTEPH with RV failure remains elusive (2). Manyattempts have been made since the 1980’s to develop a reliablemodel of CTEPH in large-animals including pigs, sheep, dogs,cows, and non-human primates with little success. Commontechniques typically include some combination of venousthrombosis, surgical ligations or shunts, balloon occlusions,embolic occlusions with microspheres or tissue adhesive, andthrombolytic or VEGF inhibitors (17–23). However, most studiesfail to either measure RV function or the model fails to replicateRV failure. Using a swine model of chronic PH (21, 22),Boulate et al. did demonstrate acute RV failure when alsoinducing volume overload via saline infusion and iterative acutepulmonary embolization (24), but the chronic PH model reliedon proximal obstruction of the right lower-lobe artery via tissue

FIGURE 1 | Experimental flowchart highlighting the major experimental

procedures.

Frontiers in Cardiovascular Medicine | www.frontiersin.org 2 January 2019 | Volume 5 | Article 189

Mulchrone et al. CTEPH: A Focus on Function

adhesive in conjunction with the PA ligation which exhibitedno clinical evidence of RV failure at rest (19). Stam et al. alsocreated a swine model of CTEPH utilizing multiple microsphereinjections in conjunction with an endothelial nitric synthaseinhibitor to cause endothelial dysfunction (25). This modeldemonstrated decreased cardiac index (CI), RV remodeling,and exercise intolerance, but required open-chested proceduresas well as a two-hit mechanism to induce hemodynamicchanges.

Here, we sought to create a canine model of CTEPH thatcould be developed using less invasive surgical procedures,as well as utilize only microsphere injections. We utilized asimilar approach as Hori et al., but followed animals untilRV failure occurred as evidenced by clinically used invasiveand non-invasive metrics such as CO, EF, and end-diastolicvolume (EDV) (26). By defining the phenotype and timing of RVfailure, we offer a clinically relevant CTEPH model as a tool forstudying the mechanism of PH-associated RV remodeling andfailure.

TABLE 1 | Estimated number of microspheres used to induce CTEPH in each

canine.

PH diagnosis Terminal end-point

Canine # Days # Microspheres # Days # Microspheres

1 116 27,000 158 29,000

2 115 36,000 199 49,000

3 238 62,000 252 62,000

4 224 61,000 252 65,000

FIGURE 2 | The progressive pressure increases in the main PA obtained from the indwelling catheter over time in a single canine that developed CTEPH.

METHODS

CTEPH was induced in five, adult male beagles (12 ± 1 kgbody weight) following a modified version of an establishedcanine model (26, 27). The protocol is outlined in Figure 1.All procedures were approved by the University of Wisconsin-Madison Institutional Animal Care and Use Committee.

Induction and AnesthesiaFollowing pre-medication with hydromorphone (0.1 mg/kg, IM)and midazolam (0.2 mg/kg, IM), general anesthesia was inducedby an intravenous (IV) injection of propofol (10 mg/kg). Theanimals were then intubated, and anesthesia was maintainedwith isoflurane (1–3%) in 100% oxygen; ventilation was adjustedas needed to maintain appropriate end-tidal CO2 levels (30–50mmHg). Sterile saline was infused via IV access at a rate of 10mL/(kg∗h). Cephalexin (30 mg/kg) was given IV. Once stableunder anesthesia, the animals were transferred to a magneticresonance imaging (MRI) suite before returning to the procedureroom.

Magnetic Resonance ImagingMRI studies were performed on a clinical 3T scanner (MR750,GE Healthcare, Waukesha, WI, USA), using previously reportedparameters (27). Briefly, axial ECG-gated CINE balanced steady-state free precession images were acquired through the entireheart. Between 20 and 30-time frames were reconstructed at eachslice location (12–25 slices depending on heart size). In addition,two-dimensional phase contrast images were acquired throughthe main, left, and right PA to assess the relative area change(RAC) and flow.

Frontiers in Cardiovascular Medicine | www.frontiersin.org 3 January 2019 | Volume 5 | Article 189

Mulchrone et al. CTEPH: A Focus on Function

Indwelling Catheter Placement andBaseline HemodynamicsCTEPH was induced by repeated injections of microspheresinto the PA via an indwelling catheter. The indwelling catheterwas inserted into the femoral vein and advanced into the PAunder fluoroscopic guidance with contrast. The distal end wastunneled under the skin and exteriorized at the dorsum betweenthe shoulder blades, where it was sutured in place to preventmovement. A Luer stub was placed on the end of the tubing anda Luer access split septum port was attached.

The femoral artery was catheterized to monitor systemicarterial pressure and arterial blood gases, and the femoraland external jugular veins were catheterized for right heartcatheterization (RHC), angiography contrast delivery, and bloodsampling. Baseline PA, RV, and right atrial (RA) pressures (mean,systolic, and diastolic) were obtained. CO was measured usingthermodilution, in triplicate. Lastly, the indwelling catheter wasfilled with heparinized saline (1000 USP Units/mL) and the endwas taped shut using self-adherent wrap.

EchocardiographyTransthoracic echocardiography was performed by a board-certified veterinary cardiologist (HBK). Animals were gentlyrestrained in lateral recumbency on a purpose-built table,with small cut-out areas under the thorax. Two-dimensionalechocardiography was used to evaluate RV size and function,while color flow and spectral Doppler imaging were performedto assess valve regurgitation. Echocardiographic measurementswere also obtained from weight- and sex-matched healthycontrols (n= 4; body weight= 11± 1 kg).

CTEPH InductionTo induce CTEPH, microspheres were perfused into thepulmonary vasculature every 3–4 days over the course ofseveral months. A 0.5mL volume of autoclaved 100–300µmmicrospheres (SephadexTM G-50 coarse; GE Healthcare) werevigorously mixed with 20mL of sterile saline and divided into2mL aliquots. Every 3–4 days, 2mL of this suspension was

FIGURE 3 | Right ventricular pressure traces from RHC at baseline and at the

terminal end-point of CTEPH.

slowly injected through the access port of the indwelling catheterfollowed by 2mL of sterile saline. The access port was replacedwith a new or disinfected port as needed. A pressure transducerconnected to the port recorded PA pressures. Then, a heparinizedsolution was added to refill the catheter. Animals were monitoredfor signs of distress (i.e., respiratory distress, shortness of breath,or collapse) before, during, and following the procedure. Toprotect catheter integrity, dogs were fitted with surgical jacketsand Elizabethan collars, and were housed individually. They werefed a commercial dry food diet and had free access to water.The access port was disinfected with 70% isopropyl alcohol andiodine, and aspirated and replaced with new heparinized salinedaily. Microsphere injections were continued until there wasevidence of hemodynamically significant PH as determined bythe veterinary cardiologist, utilizing echo tomonitor progression.Echo measurements of tricuspid regurgitation flow velocity,RV septal flattening, RV concentric hypertrophy, RV dilation,notching of the PA flow profile, and pulmonic regurgitationvelocities were performed monthly, at a minimum (28, 29).

Terminal ProcedureAs with the baseline procedure, the animals were pre-medicatedwith hydromorphone (0.1 mg/kg, IM) and midazolam (0.2mg/kg, IM), and general anesthesia was induced by an IVinjection of propofol (10mg/kg). Atropine (0.02mg/kg) was usedas needed to stabilize the heart rate. Animals were then intubated,and 0.9% saline was started IV (10 mL/(kg∗h)). Animals weretransferred to the MRI suite where RV and PA structuraland flow measurements were measured. The echocardiographicmeasurements were obtained, and then fluoroscopic guidance

TABLE 2 | Data collected from RHC and MRI before and after chronic

embolization (n = 4).

Parameter Technique Baseline CTEPH p-value

Body weight (kg) – 12 ± 1 12 ± 1 0.294

Heart rate (bpm) – 90 ± 2 108 ± 5 0.051

sPAP (mmHg) RHC 26.5 ± 3.0 44.6 ± 8.3 0.106

dPAP (mmHg) RHC 11.5 ± 1.2 26.5 ± 5.0 0.036

mPAP (mmHg) RHC 16.5 ± 1.6 34.3 ± 6.0 0.046

PCWP (mmHg) RHC 10.3 ± 0.5 10.3 ± 1.3 1.000

SBP (mmHg) RHC – 128 ± 18 –

DBP (mmHg) RHC – 72 ± 11 –

MBP (mmHg) RHC – 94 ± 13 –

RAP (mmHg) RHC 6.25 ± 0.95 7.25 ± 1.11 0.630

sRVP (mmHg) RHC 24.63 ± 3.05 43.50 ± 6.84 0.064

dRVP (mmHg) RHC 3.75 ± 1.89 5.50 ± 1.32 0.544

mRVP (mmHg) RHC 12.50 ± 1.26 20.25 ± 3.09 0.072

RV EDV (mL/kg) MRI 2.73 ± 0.06 2.98 ± 0.02 0.021

RV ESV (mL/kg) MRI 1.41 ± 0.07 2.05 ± 0.06 0.012

RV SV (mL/kg) MRI 1.32 ± 0.07 0.92 ± 0.06 0.013

RHC, right heart catheterization; MRI, magnetic resonance imaging; bpm, beats per

minute; sPAP, systolic pulmonary arterial pressure; dPAP, diastolic pulmonary arterial

pressure; mPAP, mean pulmonary arterial pressure; PCWP, pulmonary capillary wedge

pressure; SBP, systolic blood pressure; DBP, diastolic blood pressure; MBP, mean blood

pressure; RAP, right atrial pressure; RV, right ventricle; sRVP, systolic RV pressure;

dRVP, diastolic RV pressure; mRVP, mean RV pressure; EDV, end-diastolic volume; ESV,

end-systolic volume; SV, stroke volume. Bold indicates p < 0.05.

Frontiers in Cardiovascular Medicine | www.frontiersin.org 4 January 2019 | Volume 5 | Article 189

Mulchrone et al. CTEPH: A Focus on Function

was used to insert a pressure catheter where end-point PA andRV pressures were recorded. These pressure measurements wereobtained followed by CO measurements, again in triplicate.The CO was corrected for body weight to account for growth,resulting in CI. Lastly, digital subtraction angiography imageswere acquired to assess changes in lung perfusion (5).

Following the study, the animals were humanely euthanized(5mL Beuthanasia, IV) and the heart, pulmonary vasculature,and lungs were harvested and preserved for histological analysisand mechanical testing as previously reported (30).

Data AnalysisThe axial ECG-gated CINE balanced steady-state free precessionimages were used to manually contour the RV in each of the 20–30 time frames for each slice location using Segment software

TABLE 3 | Data collected during echo between CTEPH and healthy controls.

Parameter Control (n = 4) CTEPH (n = 4) p-value

Body weight (kg) 11 ± 1 12 ± 1 0.531

Heart rate (bpm) 89 ± 11 103 ± 9 0.387

Ao diameter (cm/kg) 0.16 ± 0.01 0.14 ± 0.01 0.340

PA diameter (cm/kg) 0.12 ± 0.01 0.14 ± 0.01 0.078

RV thickness (cm) 0.52 ± 0.07 0.61 ± 0.03 0.352

RV PEP (ms) 35 ± 3 42 ± 6 0.359

RV AT (ms) 87 ± 6 94 ± 13 0.612

RV ET (ms) 211 ± 16 283 ± 14 0.019

AT:ET 0.41 ± 0.02 0.33 ± 0.03 0.099

LA diameter (cm/kg) 0.20 ± 0.01 0.17 ± 0.01 0.086

LV mass (g/kg) 6.23 ± 0.55 3.89 ± 0.29 0.019

LV EDV (mL/kg) 2.55 ± 0.19 1.83 ± 0.13 0.026

LV ESV (mL/kg) 0.96 ± 0.11 0.63 ± 0.08 0.060

LV SV (mL/kg) 1.59 ± 0.20 1.20 ± 0.07 0.161

LV EF (%) 62 ± 5 66 ± 2 0.522

LVIDd (cm) 3.20 ± 0.04 2.49 ± 0.08 0.001

LVIDs (cm) 2.24 ± 0.09 1.54 ± 0.06 0.001

LVPWd (cm) 0.78 ± 0.06 0.89 ± 0.07 0.273

LVPWs (cm) 1.09 ± 0.07 1.18 ± 0.08 0.457

IVSd (cm) 0.89 ± 0.04 0.75 ± 0.04 0.041

IVSs (cm) 1.16 ± 0.08 1.01 ± 0.03 0.142

PV peak V (m/s) 0.90 ± 0.10 0.79 ± 0.09 0.449

PV gradient (mmHg) 3.4 ± 0.8 2.6 ± 0.6 0.470

PR peak V (m/s) 0 0.21 ± 0.07 0.061

PR gradient (mmHg) 0 0.2 ± 0.1 0.078

TR peak V (m/s) 0 2.64 ± 0.27 0.002

TR gradient (mmHg) 0 28.8 ± 6.2 0.019

Ao peak V (m/s) 1.04 ± 0.15 0.73 ± 0.07 0.148

Ao gradient (mmHg) 4.58 ± 1.38 2.17 ± 0.43 0.198

MV E (m/s) 0.71 ± 0.06 0.57 ± 0.03 0.121

MV A (m/s) 0.41 ± 0.07 0.37 ± 0.06 0.654

MV E/A 1.81 ± 0.20 1.71 ± 0.27 0.787

ECHO, echocardiography; bpm, beats per minute; Ao, aorta; PA, pulmonary artery; RV,

right ventricle; PEP, pre-ejection period; AT, acceleration time; ET, ejection time; LA, left

atrium; LV, left ventricle; EDV, end-diastolic volume; ESV, end-systolic volume; SV, stroke

volume; EF, ejection fraction; IDd, inner-diameter at diastole; IDs, inner-diameter at systole;

PWd, posterior wall at diastole; PWs, posterior wall at systole; IVSd, interventricular

septum thickness at diastole; IVSs, interventricular septum thickness at systole; PV,

pulmonic valve; V, velocity; PR, pulmonic regurgitation; TR, tricuspid; MV, mitral valve.

Bold indicates p < 0.05.

(Medviso, Lund, Sweden). The RV volume was calculated for allphases of the cardiac cycle, and the EDV and end-systolic volume(ESV) were taken as the maximum and minimum reconstructedvolume, respectively. Stroke volume (SV) and EF were thencalculated as:

SV = EDV− ESV (1)

EF =SV

EDV(2)

FIGURE 4 | Changes in arterial properties as described by (A) pulmonary

vascular resistance, (B) total arterial compliance, and (C) effective arterial

elastance (*p < 0.05).

Frontiers in Cardiovascular Medicine | www.frontiersin.org 5 January 2019 | Volume 5 | Article 189

Mulchrone et al. CTEPH: A Focus on Function

and a volume-only method was used to estimate the ventricular-vascular coupling (VVC) ratio (27):

VVC =SV

ESV(3)

By combining volumes with recorded pressures, the totalarterial compliance, right ventricular stroke work (RVSW), andpulmonary vascular resistance were calculated as:

Total arterial compliance =SV

sPAP− dPAP(4)

RVSW = (mPAP− RAP) ∗ SV (5)

PVR =mPAP− PCWP

CO(6)

where sPAP and dPAP are the systolic and diastolic pulmonaryarterial pressures, respectively, and RAP is the right atrial

FIGURE 5 | Changes observed in the PA. (A) The relative area change in the

MPA, LPA, and RPA before and after chronic embolization as measured from

MRI, and (B) The relative PA diameter normalized to the aortic diameter as

measured from echo (*p < 0.05).

pressure. The PA cross-sectional area and blood flow wereanalyzed using the magnitude and phase images of the two-dimensional phase contrast MRI scans, respectively. The cross-sectional area at peak systole (Amax) and end diastole (Amin) werethen used to calculate the RAC in each of the PAs (31):

RAC =Amax − Amin

Amax(7)

RAC is a non-invasive measure of proximal arterial stiffening anda predictor of mortality in PH (32). Lastly, a modified version ofthe Windkessel model was used to estimate the effective arterialelastance (Ea), a measure of RV afterload (27):

Ea =mPAP− PCWP

SV(8)

FIGURE 6 | (A) Quantification of the average flow in the MPA, LPA, and RPA

before and after chronic embolization as determined from MRI, and (B) Digital

subtraction angiography image from a canine with CTEPH; very little perfusion

in the left lung compared to the right lung (*p < 0.05).

Frontiers in Cardiovascular Medicine | www.frontiersin.org 6 January 2019 | Volume 5 | Article 189

Mulchrone et al. CTEPH: A Focus on Function

Statistical AnalysisAll data are reported as the mean ± standard error. TheRyan-Joiner test was used to check for normality. Comparisonsbetween the control and the CTEPH echocardiography datawere conducted using a two-sample t-test. Comparisons betweenbaseline and CTEPH MRI and RHC data were analyzed usinga paired t-test. A p-value< 0.05 was used to indicate statisticalsignificance. All analyses were conducted on MiniTab R© software(PA State College, version 18).

RESULTS

CTEPH InductionCTEPH was successfully induced in four of the five dogs—the data from one dog was excluded from analysis for failingto meet the requirements for PH diagnosis, mainly insufficientincreases in PA pressures. Table 1 summarizes the CTEPHinduction times for each canine as well as the estimated numberof microspheres. Figure 2 shows the progressive increases in PApressures obtained from the indwelling catheter, concurrentlyplotted with the microsphere injections for one of the caninesthat developed CTEPH, and Figure 3 shows RV pressure traces

obtained during RHC at baseline and CTEPH for the sameanimal. Table 2 summarizes the data obtained from RHC andMRI, and Table 3 contains the data from echo. Overall, chronicembolization caused an increase in PA pressures; mPAP anddPAP doubled (p = 0.046 and p = 0.036, respectively), whilesPAP increased by almost 70% (p = 0.106). The PCWP remainunchanged at 10.3± 1.3 mmHg (p= 1.0).

RV AfterloadThe chronic injection of microspheres caused an almost seven-fold increase in PVR (4.1 ± 1.1 v. 27.6 ± 5.0 [Wood units];p = 0.022) (Figure 4A). It also caused an approximately 45%reduction in the total arterial compliance (1.16 ± 0.16 v. 0.64 ±

0.07 [mmHg]; p = 0.08) (Figure 4B). The increase in PVR anddecrease in compliance resulted in a four-fold increase in Ea (0.38± 0.09 v. 2.15± 0.34 [mmHg/mL]; p= 0.012) (Figure 4C).

Proximal artery stiffening was assessed non-invasively usingthe RAC of the main, left, and right pulmonary arteries (MPA,LPA, and RPA, respectively) calculated from the MRI images.The RAC of the MPA was decreased by approximately 45% andthe RPA by approximately 26% (Figure 5A). In addition, themaximal diameter of the PA relative to the maximal diameter

FIGURE 7 | Representative MR images at end-diastole and end-systole for baseline and at the terminal end-point of CTEPH.

Frontiers in Cardiovascular Medicine | www.frontiersin.org 7 January 2019 | Volume 5 | Article 189

Mulchrone et al. CTEPH: A Focus on Function

of the aorta was significantly higher in the CTEPH animals,demonstrating PA dilation (Figure 5B).

The average blood flow was also calculated utilizing the two-dimensional phase contrast images of the PA, which decreasedthrough the MPA after chronic embolization (Figure 6A).Despite there being significantly more flow in the RPA beforeembolization, more emboli were delivered to the left lung,resulting in a significant decrease in perfusion at the end of thestudy. This was confirmed by digital subtraction angiography(Figure 6B) and necropsy.

RV FunctionRV volumes calculated from the MRI contours revealedsignificant increases in the RV end-diastolic and end-systolicvolumes (Table 2). Figure 7 shows representative MRI imagesin the same canine at end-diastole and end-systole for baselineand CTEPH measurements. These findings were supported by

FIGURE 8 | (A) Chamber volumes of the right heart from echo measurements,

and (B) severe RV dilation in a dog with severe CTEPH. Anterior view

comparing the dilated right ventricular outflow tract (RVOT) to the normal LV

(*p < 0.05).

echo measurements (Figure 8A). Dilation of the RV outflowtract was visually apparent following tissue harvest (Figure 8B).It is noteworthy that tricuspid valve vegetative endocarditis,evident by echo and at necropsy by visual inspection (Figure 9),developed in all CTEPH animals.

CTEPH also caused RV failure, as evidenced by a 40%reduction in the CI (Figure 10A), a 36% reduction inEF (Figure 10B), and 80% increase in RVSW (Figure 10C)and significant ventricular-vascular uncoupling (Figure 10D).Moreover, by echo, CTEPH animals had significantly longer RVejection times (Table 3) and decreased tricuspid annular planesystolic excursion (TAPSE) (Figure 11), which is often used asa clinical metric of RV function (33).

DISCUSSION

The lack of small and large animal models that recapitulatethe key features of clinical CTEPH has impeded progress on

FIGURE 9 | Visual inspection of the ventricles. (A) Tricuspid endocarditis. The

arrows are showing the septal and mural tricuspid valve leaflets, and (B) a

normal anterior mitral valve leaflet, shown by the arrow.

Frontiers in Cardiovascular Medicine | www.frontiersin.org 8 January 2019 | Volume 5 | Article 189

Mulchrone et al. CTEPH: A Focus on Function

FIGURE 10 | Assessment of RV function as described by (A) cardiac index,

(B) ejection fraction, (C) right ventricular stroke work, and (D)

ventricular-vascular coupling ratio (*p < 0.05).

FIGURE 11 | RV function as described by TAPSE, measured via echo (*p <

0.05).

successful, early diagnostics and greatly limited the availabletherapeutic and pharmaceutical treatment options. Manyinvestigators have attempted to create large animal models ofCTEPH utilizing various approaches with varying success, butfew have successfully captured the clinical endpoint—RV failure.In this present study, we replicated this hallmark feature ofclinical CTEPH in a canine model utilizing only the recurringinjections of microspheres.

Contrary to several other studies where microspheres alonedid not sufficiently invoke a hemodynamic or histologicalresponse (15, 25), we showed its feasibility to induce RV failurein a large animal model of CTEPH as evidenced by a significantlyreduced CI, EF, and TAPSE. The advantages of the microspheremodels are the ease of delivery, comparatively less-invasivesurgical procedures, and the ability to generate CTEPH in bothlungs. Failure of previous models is most likely due to thelack of consensus on the optimal embolic material, size, anddelivery frequency, as well as the overall time necessary toachieve functional or structural changes. While we have shownthe capability of this technique, one of the main disadvantagesis the duration necessary to achieve these results. It required 6months, on average (range 4–8 months), before sufficient levelsof PH were observed, which is much longer than most studieshave attempted (3, 18, 20, 21, 25).

An extended induction phase does have some advantage asit is notably closer to the rate of disease progression seen inhumans (10), and unlike acute studies, allows for RV remodeling.RV remodeling was determined by RV enlargement as assessedby MRI, echo, and visual inspection following the study. WhileRV remodeling can be beneficial, we speculate that these changeswere already maladaptive as systolic function had declined andVVC was shown to have decreased to approximately 0.45,signifying severe uncoupling.

While not the focus of this study, this model could offer theopportunity to study the mechanistic progression of CTEPH

Frontiers in Cardiovascular Medicine | www.frontiersin.org 9 January 2019 | Volume 5 | Article 189

Mulchrone et al. CTEPH: A Focus on Function

development. Optimization of the embolic delivery may allowfor insight into the vascular response and adaptation occurringbetween each embolization, as well as how these alterations leadto RV remodeling, the transition from adaptive to maladaptiveRV remodeling, as well as mechanistic sex differences. A betterunderstanding of this progression could lead to earlier diagnosticmarkers and better treatment options.

Digital subtraction angiography images showed that therewas an uneven distribution of microspheres delivered to theLPA compared to the RPA, which was visually confirmed atnecropsy. Since the RPA typically has more flow, we believe thisis due to the tip of the catheter being directed more towardthe LPA, causing a disproportionate number of beads to bedelivered. If the catheter tip was placed more proximally, wespeculate that there would have been a significant decrease in flowthrough the RPA between baseline and CTEPH measurements,and the flow distribution would be more equal across both lungs.Furthermore, two catheters could be utilized and positioned suchthat the microsphere distribution between the two lungs could bedelivered as desired.

Limitations of this study include the asynchronous acquisitionof pressures, volumes, and cardiac output, the variable inductiontime for CTEPH development, as well as the lack of histologicalanalysis from the RV. In addition, PH is associated with increasedanesthesia risk so several of the animals were given a dose ofatropine, an anti-muscarinic agent that directly increases heartrate by decreasing the parasympathetic tone on the sinoatrialnode (34). The half-life of atropine is relatively short, so wedo not suspect that this agent had any substantial influenceon our other metrics of interest such as RV volumes. Phasecontract MRI has been shown to underestimate flow withturbulent stenotic jets, which can cause significant signal loss dueto intravoxel dephasing (35). This is one possible explanationfor the increase in flow mismatch between RPA+LPA flowand MPA flow that occurred with the CTEPH animals. Lastly,having an indwelling catheter within the heart for that lengthof time created several challenges. One such challenge was theincreased risk of infection and sepsis. The canines were closelymonitored for fever, distress, and other signs of infection, and

given antibiotics as recommended by veterinary staff. Three ofthe four dogs also dislodged their indwelling catheter, requiringan additional invasive procedure to secure another one in placefor microsphere delivery and pressure monitoring. We alsobelieve the catheter contributed to the development of vegetativeendocarditis that not only led to tricuspid regurgitation, but alsoseverely limited catheter access during the terminal procedureand prevented our complete RHC and hemodynamic studies.Lastly, the tricuspid regurgitation, and to a lesser extent thepulmonic regurgitation, adds uncertainty to our measurement ofSV.

In conclusion, CTEPH was induced in a canine modelusing repeated injections of microspheres into the PA via anindwelling catheter, successfully inducing RV failure and RVremodeling, which has not been observed in previous acuteembolization models (27). Since the progression of pulmonaryvascular pathology to RV failure is still poorly understood, thislarge animal model could provide valuable insight into diseaseprogression. The recapitulation of heart failure phenotypes inlarge animals could provide critical links for therapeutic andpathophysiologic intervention in clinical practices, and warrantsfurther study.

AUTHOR CONTRIBUTIONS

NC, CF, TH, MB, HK, and OF designed the study. HK, TH, andOF collected the data and managed animal care. AM, HK, andNC contributed to data analysis and interpretation. AMwrote themanuscript. All authors reviewed and approved the manuscript.

FUNDING

This work was supported by the National Institutes of Health(NIH) grant 5R01HL105598 (NCC).

ACKNOWLEDGMENTS

Technical support from Daniel Consigny and KathleenHenderson is gratefully acknowledged.

REFERENCES

1. McNeil K, Dunning J. Chronic thromboembolic pulmonary hypertension

(CTEPH). Heart (2007) 93:1152–8. doi: 10.1136/hrt.2004.053603

2. Mercier O, Fadel E. Chronic thromboembolic pulmonary

hypertension: animal models. Eur Respir J. (2013) 41:1200–6.

doi: 10.1183/09031936.00101612

3. Arias-Loza PA, Jung P, Abesser M, Umbenhauer S, Williams T, Frantz S,

et al. Development and characterization of an inducible rat model of chronic

thromboembolic pulmonary hypertension. Hypertension (2016) 67:1000–5.

doi: 10.1161/HYPERTENSIONAHA.116.07247

4. Galie N, Hoeper MM, Humbert M, Torbicki A, Vachiery JL, Barbera

JA, et al. Guidelines for the diagnosis and treatment of pulmonary

hypertension: the Task Force for the Diagnosis and Treatment of Pulmonary

Hypertension of the European Society of Cardiology (ESC) and the European

Respiratory Society (ERS), endorsed by the International Society of Heart

and Lung Transplantation (ISHLT). Eur Heart J. (2009) 30:2493–537.

doi: 10.1093/eurheartj/ehp297

5. Lang IM, Madani M. Update on chronic thromboembolic

pulmonary hypertension. Circulation (2014) 130:508–18.

doi: 10.1161/CIRCULATIONAHA.114.009309

6. Fedullo PF, Auger WR, Kerr KM, Rubin LJ. Chronic thromboembolic

pulmonary hypertension. N Engl J Med. (2001) 345:1465–72.

doi: 10.1056/NEJMra010902

7. Moser KM, Bloor CM. Pulmonary vascular lesions occurring in patients with

chronic major vessel thromboembolic pulmonary hypertension. Chest (1993)

103:685–92. doi: 10.1378/chest.103.3.685

8. Ryu JH, Olson EJ, Pellikka PA. Clinical recognition of pulmonary embolism:

problem of unrecognized and asymptomatic cases. Mayo Clin Proc. (1998)

73:873–9. doi: 10.4065/73.9.873

9. Meignan M, Rosso J, Gauthier H, Brunengo F, Claudel S, Sagnard L, et al.

Systematic lung scans reveal a high frequency of silent pulmonary embolism

in patients with proximal deep venous thrombosis. Arch Intern Med. (2000)

160:159–64. doi: 10.1001/archinte.160.2.159

10. Stenmark KR, Meyrick B, Galie N, Mooi WJ, McMurtry IF. Animal models

of pulmonary arterial hypertension: the hope for etiological discovery and

Frontiers in Cardiovascular Medicine | www.frontiersin.org 10 January 2019 | Volume 5 | Article 189

Mulchrone et al. CTEPH: A Focus on Function

pharmacological cure. Am J Physiol Lung Cell Mol Physiol. (2009) 297:L1013–

32. doi: 10.1152/ajplung.00217.2009

11. O’Connell C, Montani D, Savale L, Sitbon O, Parent F, Seferian A, et al.

Chronic thromboembolic pulmonary hypertension. Presse Med. (2015)

44(12 Pt 2):e409–416. doi: 10.1016/j.lpm.2015.10.010

12. Shi W, Hu F, Kassouf W, Michel RP. Altered reactivity of pulmonary vessels

in postobstructive pulmonary vasculopathy. J Appl Physiol. (2000) 88:17–25.

doi: 10.1152/jappl.2000.88.1.17

13. Fowlkes JB, Strieter RM, Downing LJ, Brown SL, Saluja A, Salles-Cunha S,

et al. Ultrasound echogenicity in experimental venous thrombosis.Ultrasound

Med Biol. (1998) 24:1175–82. doi: 10.1016/S0301-5629(98)00089-1

14. Wagner EM, Jenkins J, Perino MG, Sukkar A, Mitzner W. Lung and vascular

function during chronic severe pulmonary ischemia. J Appl Physiol. (2011)

110:538–44. doi: 10.1152/japplphysiol.01308.2010

15. Neto-Neves EM, Brown MB, Zaretskaia MV, Rezania S, Goodwill AG,

McCarthy BP, et al. Chronic embolic pulmonary hypertension caused by

pulmonary embolism and vascular endothelial growth factor inhibition. Am J

Pathol. (2017) 187:700–12. doi: 10.1016/j.ajpath.2016.12.004

16. Dixon JA, Spinale FG. Large animal models of heart failure: a critical link

in the translation of basic science to clinical practice. Circ Heart Fail (2009)

2:262–71. doi: 10.1161/CIRCHEARTFAILURE.108.814459

17. Moser KM, Cantor JP, Olman M, Villespin I, Graif JL, Konopka R, et al.

Chronic pulmonary thromboembolism in dogs treated with tranexamic acid.

Circulation (1991) 83:1371–9. doi: 10.1161/01.CIR.83.4.1371

18. Kim H, Yung GL, Marsh JJ, Konopka RG, Pedersen CA, Chiles PG, et al.

Endothelin mediates pulmonary vascular remodelling in a canine model of

chronic embolic pulmonary hypertension. Eur Respir J. (2000) 15:640–8.

doi: 10.1034/j.1399-3003.2000.15d04.x

19. Mercier O, Tivane A, Dorfmuller P, de Perrot M, Raoux F, Decante B, et al.

Piglet model of chronic pulmonary hypertension. Pulm Circ. (2013) 3:908–15.

doi: 10.1086/674757

20. Aguero J, Ishikawa K, Fish KM, Hammoudi N, Hadri L, Garcia-Alvarez A,

et al. Combination proximal pulmonary artery coiling and distal embolization

induces chronic elevations in pulmonary artery pressure in Swine. PLoS ONE

(2015) 10:e0124526. doi: 10.1371/journal.pone.0124526

21. Boulate D, Perros F, Dorfmuller P, Arthur-Ataam J, Guihaire J, Lamrani

L, et al. Pulmonary microvascular lesions regress in reperfused chronic

thromboembolic pulmonary hypertension. J Heart Lung Transplant. (2015)

34:457–67. doi: 10.1016/j.healun.2014.07.005

22. Guihaire J, Haddad F, Noly PE, Boulate D, Decante B, Dartevelle P, et al. Right

ventricular reserve in a piglet model of chronic pulmonary hypertension. Eur

Respir J. (2015) 45:709–17. doi: 10.1183/09031936.00081314

23. Rothman A, Wiencek RG, Davidson S, Evans WN, Restrepo H, Sarukhanov

V, et al. Challenges in the development of chronic pulmonary hypertension

models in large animals. Pulm Circ. (2017) 7:156–66. doi: 10.1086/690099

24. Boulate D, Arthur Ataam J, Connolly AJ, Giraldeau G, Amsallem M,

Decante B, et al. Early development of right ventricular ischemic lesions

in a novel large animal model of acute right heart failure in chronic

thromboembolic pulmonary hypertension. J Card Fail. (2017) 23:876–86.

doi: 10.1016/j.cardfail.2017.08.447

25. Stam K, van Duin RWB, Uitterdijk A, Cai Z, Duncker DJ, Merkus D. Exercise

facilitates early recognition of cardiac and vascular remodeling in chronic

thromboembolic pulmonary hypertension in swine. Am J Physiol Heart Circ

Physiol. (2018) 314:H627–H642. doi: 10.1152/ajpheart.00380.2017

26. Hori Y, Uchide T, Saitoh R, Thoei D, Uchida M, Yoshioka K, et al.

Diagnostic utility of NT-proBNP and ANP in a canine model of

chronic embolic pulmonary hypertension. Vet J. (2012) 194:215–21.

doi: 10.1016/j.tvjl.2012.03.022

27. Bellofiore A, Roldan-Alzate A, Besse M, Kellihan HB, Consigny DW, Francois

CJ, et al. Impact of acute pulmonary embolization on arterial stiffening and

right ventricular function in dogs. Ann Biomed Eng. (2013) 41:195–204.

doi: 10.1007/s10439-012-0635-z

28. Kellihan HB, Stepien RL. Pulmonary hypertension in canine

degenerative mitral valve disease. J Vet Cardiol. (2012) 14:149–64.

doi: 10.1016/j.jvc.2012.01.001

29. Soydan LC, Kellihan HB, Bates ML, Stepien RL, Consigny DW, Bellofiore A,

et al. Accuracy of Doppler echocardiographic estimates of pulmonary artery

pressures in a canine model of pulmonary hypertension. J Vet Cardiol. (2015)

17:13–24. doi: 10.1016/j.jvc.2014.10.004

30. Golob MJ, Tabima DM, Wolf GD, Johnston JL, Forouzan O, Mulchrone

AM, et al. Pulmonary arterial strain- and remodeling-induced stiffening are

differentiated in a chronic model of pulmonary hypertension. J Biomech.

(2017) 55:92–8. doi: 10.1016/j.jbiomech.2017.02.003

31. Tian L, Kellihan HB, Henningsen J, Bellofiore A, Forouzan O, Roldan-

Alzate A, et al. Pulmonary artery relative area change is inversely

related to ex vivo measured arterial elastic modulus in the canine

model of acute pulmonary embolization. J Biomech. (2014) 47:2904–10.

doi: 10.1016/j.jbiomech.2014.07.013

32. Gan CT, Lankhaar JW, Westerhof N, Marcus JT, Becker A, Twisk

JW, et al. Noninvasively assessed pulmonary artery stiffness predicts

mortality in pulmonary arterial hypertension. Chest (2007) 132:1906–12.

doi: 10.1378/chest.07-1246

33. Ghio S, Klersy C, Magrini G, D’Armini AM, Scelsi L, Raineri C,

et al. Prognostic relevance of the echocardiographic assessment of

right ventricular function in patients with idiopathic pulmonary arterial

hypertension. Int J Cardiol. (2010) 140:272–8. doi: 10.1016/j.ijcard.2008.

11.051

34. Widmaier EP, Raff H, Strang KT, Vander AJ. Vander’s Human Physiology : The

Mechanisms of Body Function. New York, NY: McGraw-Hill (2014).

35. O’Brien KR, Cowan BR, Jain M, Stewart RA, Kerr AJ, Young AA. MRI phase

contrast velocity and flow errors in turbulent stenotic jets. J Magn Reson

Imaging (2008) 28:210–8. doi: 10.1002/jmri.21395

Conflict of Interest Statement: The authors declare that the research was

conducted in the absence of any commercial or financial relationships that could

be construed as a potential conflict of interest.

Copyright © 2019 Mulchrone, Kellihan, Forouzan, Hacker, Bates, Francois and

Chesler. This is an open-access article distributed under the terms of the Creative

Commons Attribution License (CC BY). The use, distribution or reproduction in

other forums is permitted, provided the original author(s) and the copyright owner(s)

are credited and that the original publication in this journal is cited, in accordance

with accepted academic practice. No use, distribution or reproduction is permitted

which does not comply with these terms.

Frontiers in Cardiovascular Medicine | www.frontiersin.org 11 January 2019 | Volume 5 | Article 189